Socioeconomic status affects the Oxford knee score and Short-Form 12 score following total knee replacement

Transcription

1 KNEE Socioeconomic status affects the Oxford knee score and Short-Form 12 score following total knee replacement N. D. Clement, P. J. Jenkins, D. MacDonald, Y. X. Nie, J. T. Patton, S. J. Breusch, C. R. Howie, L. C. Biant From Department of Orthopaedics and Trauma, Royal Infirmary of Edinburgh, Edinburgh, United Kingdom N. D. Clement, MRCSEd, Clinical Lecturer D. McDonald, BA, Research Assistant University of Edinburgh, Department of Orthopaedics, The Royal Infirmary of Edinburgh, Old Dalkeith Road, Edinburgh EH16 4SU, UK. P. J. Jenkins, MRSCEd, Speciality Trainee Y. X. Nie, MD, Arthroplasty Fellow J. T. Patton, FRCSEd(Tr & Orth), Consultant Orthopaedic Surgeon S. J. Breusch, FRCSEd(Tr & Orth), MD, PhD, Professor of Orthopaedics C. R. Howie, FRCSEd, Consultant Orthopaedic Surgeon L. C. Biant, FRCSEd(Tr & Orth), MS, Consultant Orthopaedic Surgeon Royal Infirmary of Edinburgh, Department of Orthopaedics and Trauma, Old Dalkeith Road, Edinburgh EH16 4SU, UK. Correspondence should be sent to Mr N. D. Clement; British Editorial Society of Bone & Joint Surgery doi: / x.95b $2.00 Bone Joint J 2013;95-B:52 8. Received 3 April 2012; Accepted after revision 12 September 2012 We assessed the effect of social deprivation upon the Oxford knee score (OKS), the Short- Form 12 (SF-12) and patient satisfaction after total knee replacement (TKR). An analysis of 966 patients undergoing primary TKR for symptomatic osteoarthritis (OA) was performed. Social deprivation was assessed using the Scottish Index of Multiple Deprivation. Those patients that were most deprived underwent surgery at an earlier age (p = 0.018), were more likely to be female (p = 0.046), to endure more comorbidities (p = 0.04) and to suffer worse pain and function according to the OKS (p < 0.001). In addition, deprivation was also associated with poor mental health (p = 0.002), which was assessed using the mental component (MCS) of the SF-12 score. Multivariable analysis was used to identify independent predictors of outcome at one year. Pre-operative OKS, SF-12 MCS, back pain, and four or more comorbidities were independent predictors of improvement in the OKS (all p < 0.001). Pre-operative OKS and improvement in the OKS were independent predictors of dissatisfaction (p = and p < 0.001, respectively). Although improvement in the OKS and dissatisfaction after TKR were not significantly associated with social deprivation per se, factors more prevalent within the most deprived groups significantly diminished their improvement in OKS and increased their rate of dissatisfaction following TKR. Cite this article: Bone Joint J 2013;95-B:52 8. Social deprivation is thought to be an aetiological factor for osteoarthritis of the knee, and according to assessments of pain and function these patients have an increased need for joint replacement. 1,2 However, this is not reflected by a higher prevalence of joint replacement surgery in this socially deprived group. 2,3 This has led some authors to call for improved access to replacement surgery for the most disadvantaged patients. 4 However, this group of patients also experiences poorer outcomes and a higher complication rate after joint replacement. 4 Socioeconomic status has been shown to affect the functional outcome of total hip replacement 5,6 but there is conflicting evidence as to whether deprivation has a bearing on the outcome of total knee replacement (TKR). Murray et al 7 found no correlation between socioeconomic status and pre-operative disease severity or the outcome after TKR. Davis et al, 8 after adjusting for confounding variables, demonstrated that disease severity pre-operatively was greater in those with the lowest financial income but they achieved the same final functional outcome as those with a higher income. In addition, more recently Lopez-Olivo et al 9 demonstrated that both multiple psychological and social factors affect the outcome of TKR. In a survey of 1217 consecutive TKRs at one year, only four of five patients undergoing TKR were found to be satisfied or very satisfied with their outcome, with the remainder being unsure or dissatisfied. 10 Multiple factors have been demonstrated to predict patient satisfaction after TKR, including patient demographics, diagnosis, disease severity and comorbidity The relationship between social deprivation and patient satisfaction after TKR has not yet been clarified, and if functional outcome is negatively influenced by deprivation, then this may in turn affect the rate of patient satisfaction. Patient-reported outcome measures (PROMs) have been introduced to the National Health Service (NHS) in the United Kingdom for audit purposes. 13 If the socioeconomic status influences PROMs, it should be acknowledged and the outcome measure adjusted to enable a fair comparison when comparing patient cohorts. The Oxford knee score (OKS) is a widely accepted PROM, 14 but no study to date has assessed the effect of socioeconomic status upon the OKS after TKR. 52 THE BONE & JOINT JOURNAL

2 SOCIOECONOMIC STATUS AFFECTS THE OXFORD KNEE SCORE AND SHORT-FORM 12 SCORE FOLLOWING TOTAL KNEE REPLACEMENT 53 Table I. Patient demographics and pre-operative functional scores according to deprivation quintile. Statistically significant p-values are given in bold Socioeconomic deprivation quintile Demographic 1 (n = 112) 2 (n = 207) 3 (n = 183) 4 (n = 182) 5 (n = 282) p-value * Gender (n, %) Male 41 (36.6) 82 (39.6) 79 (43.2) 94 (51.6) 125 (44.3) Female 71 (63.4) 125 (60.4) 104 (56.8) 88 (48.4) 157 (55.7) Mean (SD) age (yrs) 69.4 (6.9) 70.4 (7.0) 69.7 (7.0) 70.0 (7.0) 72.1 (7.2) Comorbidity (n, %) Heart disease 25 (22.3) 25 (12.1) 21 (11.5) 33 (18.1) 38 (13.5) 0.04 Hypertension 51 (45.5) 87 (42.0) 75 (41.0) 81 (44.5) 113 (40.0) 0.82 Lung disease 13 (11.6) 23 (11.1) 23 (12.6) 20 (11.0) 18 (6.4) 0.18 Vascular disease 5 (4.5) 11 (5.3) 6 (3.3) 2 (1.1) 12 (4.3) 0.24 Neurological disease 5 (4.5) 4 (1.9) 3 (0.5) 2 (1.1) 13 (4.6) 0.09 Diabetes mellitus 12 (10.7) 18 (8.7) 17 (9.3) 16 (8.8) 33 (11.7) 0.78 Gastric ulceration 3 (2.7) 8 (3.9) 8 (4.4) 7 (3.8) 15 (5.3) 0.8 Kidney disease 0 3 (1.4) 4 (2.2) 1 (0.5) 3 (1.1) 0.43 Liver disease 0 3 (1.4) 4 (2.2) 1 (0.5) 1 (0.4) 0.19 Anaemia 9 (8.0) 10 (4.8) 6 (3.3) 3 (1.6) 15 (5.3) 0.1 Back pain 40 (35.7) 74 (35.7) 64 (35.0) 66 (36.2) 99 (35.1) 0.9 Depression 15 (13.4) 24 (11.6) 13 (7.1) 15 (8.2) 18 (6.4) comorbidities (n, %) 14 (12.5) 18 (8.7) 9 (4.9) 7 (3.8) 17 (6.0) 0.04 Mean (SD) pre-operative scores Oxford knee score 43.3 (6.5) 42.6 (7.1) 41.3 (7.3) 40.6 (7.7) 39.5 (7.5) < SF-12 PCS 30.1 (9.7) 30.1 (8.7) 30.7 (9.5) 30.4 (10.3) 30.0 (8.1) 0.94 SF-12 MCS 48.0 (12.0) 49.1 (12.1) 51.5 (11.3) 51.8 (12.7) 52.2 (11.3) * chi-squared test, unless otherwise stated analysis of variance SF-12, Short-Form 12; PCS, physical component score; MCS, mental health component scores Patients and Methods Over a three-year period (2007 to 2009) 996 patients with primary osteoarthritis (OA) who underwent TKR were identified. All patients attended a pre-operative assessment, at which any comorbidities (including heart disease, hypertension, lung disease, vascular disease, neurological problems, stomach ulcer, kidney disease, liver disease, depression and back pain) were recorded as dichotomous variables. The OKS and the Short-Form (SF)-12 scores 15 were recorded pre-operatively and at one year postoperatively. The OKS consists of twelve questions assessed on a Likert scale with values from 1 to 5, a summative score is then calculated where 12 is the best possible score (least symptomatic) and 60 is the worst possible score (most symptomatic). The series comprised 421 men (mean age 70.7 years (37 to 91; SD 8.9)) and 545 women (mean age 70.5 years (33 to 93; SD 9.2)). Of these, 290 patients (30%) did not have a medical comorbidity, but 65 (6.7%) had four comorbidities. The mean pre-operative OKS was 41.1 (SD 7.4), mean SF-12 physical component score (PCS) was 30.3 (SD 9.1) and mean SF-12 mental component score (MCS) was 50.9 (SD 11.9) (Table I). Patient satisfaction was assessed with the patient selecting one of four possible responses one year after surgery: very satisfied, satisfied, unsure and unsatisfied. This has been used previously to assess patient satisfaction after TKR. 10 The Scottish Index of Multiple Deprivation (SIMD), 16 which takes into account employment, income, crime, housing, health, education and access to services, was used to assess the patients socioeconomic status. Patients were allocated to a social quintile according to their postcode, which ranged from deprivation quintile 1 (most deprived) to quintile 5 (least deprived). 17 During the study period the most commonly performed TKRs were the Kinemax (Stryker Howmedica Osteonics, Allendale, New Jersey), Triathlon (Stryker), and the PFC Sigma (DePuy, Johnson & Johnson Professional Inc., Raynham, Massachusetts). A standardised rehabilitation protocol was used for all patients, with active mobilisation on the first post-operative day. Patients were then reviewed at six weeks, six months and 12 months post-operatively. The length of stay of each patient was recorded. Ethical approval was obtained for analysis and publication of the presented data from the regional ethics committee. Statistical analysis. Statistical analysis was performed using Statistical Package for Social Sciences v17.0 (SPSS Inc., Chicago, Illinois). Parametric and non-parametric tests were used as appropriate to assess continuous variables for significant differences between groups. A Student s t-test, both unpaired and paired, a one-way analysis of VOL. 95-B, No. 1, JANUARY 2013

3 54 N. D. CLEMENT, P. J. JENKINS, D. MACDONALD, Y. X. NIE, J. T. PATTON, S. J. BREUSCH, C. R. HOWIE, L. C. BIANT Pre-operative OKS Pre-operative SF-12 MCS Fig. 1 R 2 = Scatter plot showing the correlation between pre-operative Short-Form (SF)-12 mental component score (MCS) and the pre-operative Oxford knee score (OKS) (r = 0.38, p < 0.001). variance (ANOVA), and a Kruskal-Wallis test were used to compare linear variables between groups, and Pearson s correlation was used to assess the relationship between linear variables. Dichotomous variables were assessed using a chi-squared or Fisher s exact test if one variable was < 10. The pre-operative OKSs were stratified into quintiles (12 to 34, 35 to 40, 41 to 45, 46 to 50 and 51 to 60) to allow for categorical analysis. Multivariate linear and bivariate regression analyses were used to identify independent predictors of outcome (improvement in OKS and patient dissatisfaction, respectively). Results There was no significant difference in the use of the three TKR implants between social quintiles (chi-squared, p = 0.9). The median length of stay for the whole cohort was five days (1 to 29, SD 3.0). There was no significant difference in length of stay between social quintiles (Kruskal-Wallis, p = 0.62). A total of 11 patients (1.1%) suffered deep infection and 13 (1.3%) had a lower-limb deep-vein thrombosis diagnosed by ultrasound (four had a non-fatal pulmonary embolism). There was no significant difference in deep infection or DVT post-operatively according to social quintile (chi-squared, p = 0.81 and p = 0.76, respectively). The pre-operative patient demographics, OKS and the components of the SF-12 score according to the SIMD quintile are shown in Table I. The most deprived patients (SIMD 1) were significantly more likely to be female (p = 0.046), younger (p = 0.018), and to have a greater level of comorbidity (p = 0.04), with heart disease being the most prevalent at 22.3% (n = 25). The most deprived suffered significantly greater pain and functional deficit according to the OKS (p < 0.001), but no significant difference was observed in the SF-12 PCS (p = 0.94). In addition, the SF-12 MCS revealed that the most deprived had a significantly less positive mental attitude (p = 0.002). The pre-operative SF-12 MCS correlated significantly with pre-operative OKS (Fig. 1) (r = 0.38, p < 0.001). Across all the TKRs the improvement in the mean OKS and mean SF-12 PCS were significant (paired t-test, p < 0.001) (Table II). However, there was no significant improvement observed for the mean MCS (paired t-test, p = 0.1). Although there were significant differences observed for the post-operative scores the absolute improvement of scores between the social quintiles was not significantly different (all p 0.32) (Fig. 2). Although there was no difference in the improvement of the OKS between social quintiles, when analysing the improvement of each pre-operative OKS group (Fig. 3), a significant difference was observed between social quintiles for those within the worst (51 to 60) pre-operative OKS group (ANOVA, p = 0.008). Pre-operative predictors of change in the OKS after TKR are illustrated in Table III. Male gender, a history of heart disease or back pain or depression, four or more comorbidities, a better pre-operative OKS (Fig. 4) or SF- 12 PCS, and a worse pre-operative SF-12 MCS significantly diminished the improvement in the OKS (Table III). These predictors were entered into multivariate regression analysis with social status (SIMD quintile), which confirmed the significant and independent effect of: back pain, four or more comorbidities, the pre-operative OKS and SF-12 MCS. Social deprivation as an isolated variable was not a significant (p = 0.068) predictor of change in the OKS (Table IV). There were 819 satisfied or very satisfied patients and 47 (4.9%) who were not satisfied with their TKR. An additional 100 were unsure whether they were satisfied or not. There was a significant difference in satisfaction between social quintiles (Table II), with the more deprived patients being less satisfied. Those patients who were unsatisfied (n = 47) were more likely to be younger (mean 67.6 years versus 70.7 years, difference 3.2 years (95% CI 0.5 to 5.8); unpaired t-test, p = 0.02), more deprived (chi-squared, p = 0.029), with a less symptomatic pre-operative OKS (mean 40.7 versus 43.7, difference 3.0 (95% CI 0.8 to 5.2); unpaired t-test, p = 0.006) and a worse pre-operative SF-12 MCS (mean 47.1 versus 51.4, difference 4.1 (95% CI 0.8 to 7.5); unpaired t-test p = 0.009), and had less of an improvement in their OKS (mean 1.5 versus 17.6, difference 16.1 (95% CI 13.5 to 18.7); unpaired t-test p < 0.001). However, gender (chi-squared, p = 0.27) and pre-operative SF- 12 PCS (unpaired t-test, p = 0.79) did not affect satisfaction. These variables were entered into bivariate regression analysis using forward stepwise methodology. This revealed pre-operative OKS and improvement in the OKS after TKR to be independent predictors of satisfaction after TKR (Table V). THE BONE & JOINT JOURNAL

4 SOCIOECONOMIC STATUS AFFECTS THE OXFORD KNEE SCORE AND SHORT-FORM 12 SCORE FOLLOWING TOTAL KNEE REPLACEMENT 55 Table II. Post-operative outcome measures and the difference relative to pre-operative scores for all patients and according to their Scottish Index of Multiple Deprivation (CI, confidence interval). Statistically significant p-values are given in bold Socioeconomic deprivation quintile Mean (SD) score * All patients (n = 966) 1 (n = 112) 2 (n = 207) 3 (n = 183) 4 (n = 182) 5 (n = 282) p-value Oxford knee score Post-operative 25.5 (10.1) 27.2 (10.4) 27.2 (10.4) 26.0 (10.3) 24.8 (10.5) 23.6 (9.0) Mean difference (95% CI) 15.7 (15.1 to 16.2) 16.1 (14.1 to 18.1) 15.4 (14.0 to 16.8) 15.3 (13.7 to 16.8) 15.9 (14.5 to 17.3) 15.9 (14.8 to 16.9) 0.92 SF-12 PCS Post-operative 40.6 (11.5) 40.2 (12.2) 40.3 (13.4) 39.8 (10.3) 40.3 (11.8) 40.9 (10.2) 0.71 Mean difference (95% CI) 10.3 (9.5 to 11.1) 10.0 (7.2 to 12.8) 10.2 (8.3 to 12.1) 9.1 (7.3 to 10.8) 10.9 (8.8 to 13.1) 10.9 (9.6 to 12.2) 0.61 SF-12 MCS Post-operative 51.5 (11.3) 49.5 (12.4) 50.2 (13.3) 51.9 (10.3) 50.8 (11.7) 51.5 (11.3) Mean difference (95% CI) 0.7 (-0.1 to 1.4) 1.5 (-1.1 to 4.0) 1.1 (-0.8 to 3.1) 0.4 (-1.2 to 2.0) -1.0 (-2.7 to 0.8) 1.2 (0.0 to 2.5) 0.32 Satisfied (n, %) Unsatisfied (n, %) 47 4 (4.0) 16 (8.8) 14 (8.4) 5 (3.1) 8 (3.2) * SF-12, Short-Form 12; PCS, physical component score; MCS, mental component score Kruskal-Wallis test analysis of variance chi-squared test OKS Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Change in OKS SIMD quintile SIMD 1 SIMD 2 SIMD 3 SIMD 4 SIMD Pre-operative Post-operative Quintiles of pre-operative OKS Fig. 2 Graph showing the pre- and post-operative mean Oxford knee score (OKS) according to social quintile (1 = most deprived, 5 = least deprived). Fig. 3 Bar chart showing the mean improvement in Oxford knee score (OKS) at 12 months according to pre-operative OKS quintile and social quintile (1 most deprived, 5 least deprived). SIMD, Scottish Index of Multiple Deprivation. Discussion Socially deprived patients undergoing TKR are significantly more likely to be female, younger, to suffer a greater level of comorbidity and to have a poorer mental health status. They also have a significantly greater level of functional deficit according to the OKS, which persists post-operatively. Although social status is not an isolated predictor of outcome, factors that are more prevalent within the more socially deprived group significantly affect the improvement of their OKS. However, once adjusting for predictors of dissatisfaction that are more prevalent in the socially deprived, there is no difference in the rate of dissatisfaction with TKR one year after surgery according to social status. A limitation to our study was the failure to record body mass index (BMI) and smoking status. In retrospect, these variables could have been assessed for their independent affect outcome, adjusting for social deprivation. Increasing BMI is associated with earlier onset of osteoarthritis and need for TKR. 18 Obesity is associated with increasing social VOL. 95-B, No. 1, JANUARY 2013

5 56 N. D. CLEMENT, P. J. JENKINS, D. MACDONALD, Y. X. NIE, J. T. PATTON, S. J. BREUSCH, C. R. HOWIE, L. C. BIANT Table III. Pre-operative predictors of change in Oxford knee score (OKS) and Short-Form (SF)-12, physical (PCS) and mental health components (MCS) after total knee replacement with univariate and bivariate analysis Demographic * Description Mean (SD) change in OKS Correlation (r) p-value Gender Male 14.9 (10.1) Female 16.3 (9.5) Mean age Comorbidities Heart disease Yes 14.2 (10.0) 0.05 No 15.9 (9.7) Hypertension Yes 15.7 (9.6) 0.99 No 15.7 (9.9) Lung disease Yes 15.9 (10.3) 0.78 No 15.7 (9.7) Vascular disease Yes 13.4 (9.9) 0.16 No 15.8 (9.8) Neurological disease Yes 14.6 (11.3) 0.56 No 15.7 (9.7) Diabetes mellitus Yes 16.4 (9.9) 0.47 No 15.6 (9.8) Gastric ulceration Yes 16.7 (9.5) 0.5 No 15.6 (9.8) Kidney disease Yes 14.8 (8.8) 0.77 No 15.7 (9.8) Liver disease Yes 13.6 (14.6) 0.51 No 15.7 (9.7) Anaemia Yes 14.0 (9.1) 0.23 No 15.8 (9.8) Back pain Yes 14.1 (10.1) < No 16.5 (9.5) Depression Yes 12.6 (9.4) No 16.0 (9.8) 4 comorbidities Yes 12.1 (9.6) No 15.9 (9.8) Socioeconomic deprivation (SIMD) Quintile (10.5) 0.93 Quintile (10.2) Quintile (10.5) Quintile (9.5) Quintile (8.8) Pre-operative score OKS 0.34 < SF-12 PCS < SF-12 MCS * SIMD, Scottish Index of Multiple Deprivation; OKS, Oxford knee score; SF-12 Short-Form 12; PCS, physical component score; MCS, mental component score Student s t-test Pearson s correlation analysis of variance deprivation, 19 and this may explain the observed younger age at time of surgery in the most deprived group. Cigarette smoking also affects the outcome of TKR 20 and is more prevalent in the most socially deprived, 21 which could have contributed to their poorer outcome. Furthermore, we did not assess the severity of the OA radiologically, which has been shown to affect the outcome of TKR, with observationally less severe degrees of OA having an inferior outcome. 22 Socially deprived patients who suffer poorer mental health, which can impair pain coping mechanisms, 23 may present at an earlier age with less severe disease, and this may have also influenced their outcome. Of the most deprived social quintile undergoing TKR, 63% were female, which is greater than the 51% prevalence observed in the same quintile for the whole population at risk. Steel et al 2 described a 30% greater need for TKR in women than in men, and in addition the poorest THE BONE & JOINT JOURNAL

6 SOCIOECONOMIC STATUS AFFECTS THE OXFORD KNEE SCORE AND SHORT-FORM 12 SCORE FOLLOWING TOTAL KNEE REPLACEMENT 57 Change in OKS Table IV. Multivariate linear regression analysis to identify pre-operative independent predictors of improvement in the Oxford knee score (OKS) after total knee replacement. All significant variables (Table II) were all entered into the model using a stepwise methodology (r 2 = 0.18) Predictor * B (95% CI) p-value Back pain (-3.75 to -1.30) < comorbidities (-6.11 to -1.45) < Pre-operative OKS 0.58 (0.50 to 0.67) < Pre-operative SF-12 MCS 0.16 (0.11 to 0.22) < Constant ( to 10.40) < * SF-12 MCS, Short-Form 12 mental component score -10 Table V. Predictors of dissatisfaction after total knee replacement on bivariate regression analysis (Nagelkerke r 2 = 0.50) (OKS, Oxford knee score; CI, confidence interval) -20 Pre-operative OKS Fig. 4 Scatter plot showing the correlation between pre-operative and improvement in the Oxford knee score (OKS) after total knee replacement (r = 0.34, p < 0.001). quintile experienced three times as much need for surgery as the wealthiest quintile. This is also supported by the significantly younger age at the time of surgery observed in the most deprived quintile. The increased level of comorbidity associated with the most deprived may be associated with poor risk behaviours, such as smoking, inappropriate diet and low levels of activity. 24 Female gender, 25 younger age 26 and increasing levels of comorbidity 27 are all associated with a poor outcome after TKR. Hence a poorer outcome in the most deprived patients enduring TKR, in whom these factors are more prevalent, might be expected. Social deprivation in our cohort was associated with a significantly worse pre-operative pain and functional status according to the OKS. Davis et al 8 demonstrated a similar observation using the Western Ontario and McMaster Universities osteoarthritis index (WOMAC), 28 where patients with a lower income had a worse pre-operative score. However, the four-point difference in OKS that we observed between the most and least deprived may not be clinically important, as the smallest change in the score to be of importance is accepted to be approximately half the standard deviation for a scoring measure. 29,30 The SD of the OKS in this series was 10. The most socially deprived of our patients had a significantly reduced mental wellbeing score according to the SF- 12 MCS. In another series, 31 patients waiting for TKR with a lower income and educational level were reported to suffer greater pre-operative psychological distress. The mental component of the SF-12 has been shown to correlate with both outcome score and patient satisfaction. 32,33 This study demonstrates a significant correlation between poor pre-operative mental health with a poor pre-operative OKS Predictor B Odds ratio (95% CI) p-value Pre-operative OKS (1.06 to 1.31) Improvement in OKS (0.70 to 0.87) < Constant and diminished improvement in the OKS post-operatively. The SF-12 MCS also correlates significantly with patient satisfaction on univariate analysis, but after taking other predictors into account on multivariate analysis it was not significant. However, the two predictors of satisfaction, pre-operative OKS and improvement in the OKS were directly related to the SF-12 MCS. Pre-operative pain and functional status have been shown to be the strongest predictors of outcome previously. 34,35 In this study we have identified a number of independent predictors of poor outcome when measured by OKS after TKR, including patients with back pain or four or more comorbidities, a better pre-operative OKS, and a worse pre-operative SF-12 MCS. A poor mental status, as measured by the SF-36, has also been demonstrated to be associated with a poorer outcome after TKR, 36 although there is also additional evidence from smaller cohorts that fails to substantiate this association. 37 Back pain as a predictor of poor outcome is unique to this study but it has previously been identified as a significant predictor of patient dissatisfaction. 10 Brinker et al 27 reported that increasing levels of comorbidity assessed by two or more major medical conditions has a significant negative correlation with outcome scores. The presence of back pain and comorbidity alone would reduce the improvement in OKS by more than six points, which is clinically significant. Using all four significant independent predictors, it would be possible pre-operatively to estimate a patient s improvement after a TKR by the following formula, derived from Table IV: Improvement in OKS = (pre-operative OKS 0.58) + (pre-operative SF-12 MCS 0.16) 2.53 if back pain 3.78 if four comorbidities 15.4 (constant). VOL. 95-B, No. 1, JANUARY 2013

7 58 N. D. CLEMENT, P. J. JENKINS, D. MACDONALD, Y. X. NIE, J. T. PATTON, S. J. BREUSCH, C. R. HOWIE, L. C. BIANT Dissatisfaction after TKR was significantly associated with social deprivation, when this is related to factors more prevalent within the most deprived patients but not with socioeconomic status per se. In addition, socially deprived patients undergoing TKR have a greater level of comorbidity and poorer mental wellbeing. These may contribute to the subjectively worse pain and functional scores observed in the most deprived. These factors should be acknowledged pre-operatively when assessing and consenting socially deprived patients for TKR. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1. Dawson J, Linsell L, Zondervan K, et al. Epidemiology of hip and knee pain and its impact on overall health status in older patients. Rheumatology (Oxford) 2004;43: Steel N, Melzer D, Gardener E, McWilliams B. Need for and receipt of hip and knee replacement - a national population survey. Rheumatology (Oxford) 2006;45: Dixon T, Shaw M, Ebrahim S, Dieppe P. Trends in hip and knee joint replacement: socioeconomic inequalities and projections of need. Ann Rheum Dis 2004;63: Ellis HB, Howard KJ, Khaleel M. Total joint replacement: influence of socioeconomic status on outcome of joint replacement surgery. Curr Orthopaed Pract 2010;21: Clement ND, Muzammil A, MacDonald D, Howie CR, Biant LC. Socioeconomic status affects the early outcome of total hip replacement. J Bone Joint Surg [Br] 2011;93-B: Jenkins PJ, Perry PRW, Ng CY, Ballantyne JA. Deprivation influences the functional outcome from total hip arthroplasty. Surgeon 2009;7: Murray JR, Birdsall PD, Sher JL, Deehan DJ. Deprivation and outcome of total knee replacement. Knee 2006;13: Davis ET, Lingard EA, Schemitsch EH, Waddell JP. Effects of socioeconomic status on patients' outcome after total knee arthroplasty. Int J Qual Health Care 2008;20: Lopez-Olivo MA, Landon GC, Siff SJ, et al. Psychosocial determinants of outcomes in knee replacement. Ann Rheum Dis 2011;70: Scott CE, Howie CR, MacDonald D, Biant LC. Predicting patient dissatisfaction following total knee replacement: a prospective study of 1217 cases. J Bone Joint Surg [Br] 2010;92-B: Kennedy DM, Hanna SE, Stratford PW, Wessel J, Gollish JD. Preoperative function and gender predict pattern of functional recovery after hip and knee arthroplasty. J Arthroplasty 2006;21: Cushnaghan J, Bennett J, Reading I, et al. Long-term outcome following total knee arthroplasty: a controlled longitudinal study. Ann Rheum Dis 2009;68: No authors listed. Department for health: equity and excellence: liberating the NHS PolicyAndGuidance/DH_ (date last accessed 12 September 2012). 14. Dawson J, Fitzpatrick R, Murray D, Carr A. Questionnaire on the perceptions of patients about total knee replacement. J Bone Joint Surg [Br] 1998;80-B: Ware J Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34: No authors listed. The Scottish Government Scottish Index of Multiple Deprivation 2009: General Report, Publications (date last accessed 12 September 2012). 17. No authors listed. The Scottish Government: Scottish Index of Multiple Deprivation Postcode to Datazone Lookup Table (date last accessed 12 September 2012). 18. Gandhi R, Wasserstein D, Razak F, Davey JR, Mahomed NN. BMI independently predicts younger age at hip and knee replacement. Obesity (Silver Spring) 2010;18: Kinra S, Nelder RP, Lewendon GJ. Deprivation and childhood obesity: a cross sectional study of 20,973 children in Plymouth, United Kingdom. J Epidemiol Community Health 2000;54: Singh JA. Smoking and outcomes after knee and hip arthroplasty: a systematic review. J Rheumatol 2011;38: Duncan C, Jones K, Moon G. Smoking and deprivation: are there neighbourhood effects? Soc Sci Med 1999;48: Merle-Vincent F, Couris CM, Schott AM, et al. Factors predicting patient satisfaction 2 years after total knee arthroplasty for osteoarthritis. Joint Bone Spine 2011;78: Verra ML, Angst F, Staal JB, et al. Differences in pain, function and coping in Multidimensional Pain Inventory subgroups of chronic back pain: a one-group pretestposttest study. BMC Musculoskelet Disord 2011;12: Macleod U, Mitchell E, Black M, Spence G. Comorbidity and socioeconomic deprivation: an observational study of the prevalence of comorbidity in general practice. Eur J Gen Pract 2004;10: Ethgen O, Bruyère O, Richy F, Dardennes C, Reginster JY. Health-related quality of life in total hip and total knee arthroplasty: a qualitative and systematic review of the literature. J Bone Joint Surg [Am] 2004;86-A: Baker PN, van der Meulen JH, Lewsey J, Gregg PJ. The role of pain and function in determining patient satisfaction after total knee replacement. Data from the National Joint Registry for England and Wales. J Bone Joint Surg [Br] 2007;89-B: Brinker MR, Lund PJ, Barrack RL. Demographic biases of scoring instruments for the results of total knee arthroplasty. J Bone Joint Surg [Am] 1997;79-A: Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically-important patient-relevant outcomes following total hip or knee arthroplasty in osteoarthritis. J Orthop Rheumatol 1998;1: Norman GR, Sloan JA, Wywich KW. Interpretatation of the changes in healthrelated quality of life: the remarkable universality of half a standard deviation. Med Care 2003;41: Schmitt JS, Di Fabio RP. Reliable change and minimum difference (MID) proportions facilitated group responsivemess comparisons using indvidual threshold criteria. J Clin Epidemiol 2004;57: Ackerman IN, Graves SE, Wicks IP, Bennell KL, Osborne RH. Severely compromised quality of life in women and those of lower socioeconomic status waiting for joint replacement surgery. Arthritis Rheum 2005;53: Walton MJ, Newman JH. Pre-operative mental wellbeing and the outcome of knee replacement. Knee 2008;15: Gandhi R, Davey JR, Mahomed NN. Predicting patient dissatisfaction following joint replacement surgery. J Rheumatol 2008;35: Fortin PR, Penrod JR, Clarke AE, et al. Timing of total joint replacement affects clinical outcomes among patients with osteoarthritis of the hip or knee. Arthritis Rheum 2002;46: Lingard EA, Katz JN, Wright EA, Sledge CB. Predicting the outcome of total knee arthroplasty. J Bone Joint Surg [Am] 2004;86-A: Ayers DC, Franklin PD, Ploutz-Snyder R, Boisvert CB. Total knee replacement outcome and coexisting physical and emotional illness. Clin Orthop Relat Res 2005;440: Fisher DA, Dierckman B, Watts MR, Davis K. Looks good but feels bad: factors that contribute to poor results after total knee arthroplasty. J Arthroplasty 2007;22: THE BONE & JOINT JOURNAL